INGERSON LECTURE The composition of the continental crust*

Geochimicaet CosmochimicaActa, Vol. 59, No. 7, pp. 1217-1232, 1995
Copyright © 1995 ElsevierScienceLtd
Printed in the USA. All rights reserved
0016-7037/95 $9.50 + .00
Pergamon
0016-7037(95)00038-0
INGERSON LECTURE
The composition of the continental crust*
K. HANS WEDEPOHL
Geochemisches Institut, D-37077 G6ttingen, Germany
A b s t r a c t - - A new calculation of the crustal composition is based on the proportions of upper crust ( U C )
to felsic lower crust ( F L C ) to mafic lower crust ( M L C ) of about 1:0.6:0.4. These proportions are derived
from a 3000 km long refraction seismic profile through western Europe ( E G T ) comprising 60% old shield
and 40% younger fold belt area with about 40 km average Moho depth. A granodioritic bulk composition
of the UC in major elements and thirty-two minor and trace elements was calculated from the Canadian
Shield data (Shaw et al., 1967, 1976). The computed abundance of thirty-three additional trace elements
in the UC is based on the following proportions of major rock units derived from mapping: 14% sedimentary rocks, 25% granites, 20% granodiorites, 5% tonalites, 6% gabbros, and 30% gneisses and mica
schists. The composition of FLC and MLC in major and thirty-six minor and trace elements is calculated
from data on felsic granulite terrains and mafic xenoliths, respectively, compiled by Rudnick and Presper
(1990). More than thirty additional trace element abundances in FLC and MLC were computed or estimated from literature data.
The bulk continental crust has a tonalitic and not a dioritic composition with distinctly higher concentrations of incompatible elements including the heat producing isotopes in our calculation. A dioritic bulk
crust was suggested by Taylor and McLennan ( 1985 ). The amount of tonalite in the crust requires partial
melting of mafic rocks with about 100 km thickness (compared with about 7 km in the present M L C ) and
water supply from dehydrated slabs and mafic intrusions. At the relatively low temperatures of old crustal
segments M L C was partly converted into eclogite which could be recycled into the upper mantle under
favourable tectonic conditions. The chemical fractionation of UC against FLC + M L C was caused by
granitoidal partial melts and by mantle degassing which has controlled weathering and accumulation of
volatile compounds close to the Earth's surface.
FORMER ESTIMATES OF CRUSTAL COMPOSITION;
SEISMIC CRUSTAL MODEL
the geometry of the crust. EGT was long and variable enough
to be representative of a typical crustal section. It comprises
the Archean and Proterozoic Fennoscandian Shield with 45.5
km average Moho depth, the Late Proterozoic and Phanerozoic fold belts of Central Europe with 30 km crustal thickness,
and the young Alpine orogen with its deep roots from the
collision of two continents. The proportion of 60% old shield
and 40% younger fold belts is representative of the state of
the continental crust in different continents. A combination of
the two crustal sections of different age gives an average
crustal thickness of 40 km.
A traveltime interpretation of the seismic refraction profile
through the Fennoscandian Shield ( F E N N O L O R A ) , the transition from the Svecofennian Province in the Sorgenfrei-Tornquist Zone to the Danish and North German Basins
( E U G E N O S) and the younger fold belts in Central Europe
( E U G E M I and EGT-S 86) is presented in Fig. 1. At the right
edge of the lower part of this cross section, Proterozoic and
Phanerozoic Central Europe merges into the deep Alpine orogen in which the European and African crusts collided. For
better presentation the simplified profile was divided into two
parts.
In Fig. 1 the continental crust consists of two major layers.
The upper and lower crust can be conventionally separated
according to P-wave velocities of less and more than 6.5 km/s.
The interface occurs at almost 20 km depth in the upper part
of Fig. 1 and is more variable in depth ( 2 0 - 5 0 km) in the
The general acceptance of the definition of the continental
crust as the continental upper part of the silicate Earth above
the Mohorovici~ discontinuity dates back to the fifties of this
century ( Poldervaart, 1955 ). Earlier authors as Clarke (1924)
and Goldschmidt (1933) reported average chemical compositions of magmatic rocks or of what the latter author called
the "lithosphere." Both averages were restricted to major
units of the upper continental crust of our present definition.
In this investigation we conclude from seismic properties
on the thickness of the continental crust and the size and compositional characteristics of its major units. During the recent
one and a half decade, refraction seismic ray velocities have
been evaluated on a 4800 km long almost continuous traverse
through Europe from the North Cape to Tunisia of which we
used --3000 km (Ansorge et al., 1992). The European Geotraverse ( E G T ) was designed to encompass the succession of
tectonic provinces from the oldest known Archean to those
still active today. The tectonic provinces occur in succession
from the north to the south as well as in time. The first step
in understanding how the continental lithosphere of Europe
formed was a survey of physical properties (density, etc.) in
* Ingerson Lecture presented during the Goldschmidt Conference
in Edinburgh, Scot August 31, 1994. The lecture was sponsored by
the International Association of Geochemistry and Cosmochemistry.
1217
1218
K.H. Wedepohl
Archean Crustal
Province
Scandinavian
Caledonides,
[ - - N ~ Barents Sea
0
Depth-
Svecofennian Crustal
Province
2040(km) 60-
I
200
I
400
Sorgenfrei-Tornquist
Zone, Baltic Sea
I
600
I
800
I
1200
I
1000
Variscian
Province
North German
Basin
I
1400
km
Alps
~~ 0Depth
2O
4O
(km)
:::!i:~i~::i~ii~iii~iiit::Ii::
ii~::iii::i::iiiiii
i!iiii!iiiiiiiiiii~
~
1600
~
1800
Lithospheric Mantle
(Vp > 6.1 km/s)
2000
2200
Mafic Granulites
(Vp 6.9 - 7.5 km/s)
2400
2600
I
2800 km
Sediments,
Granites,
Gneisses
Felsic Granulites
(Vp 6.5 - 6.9 km/s)
(Vp < 6 - 6.5 km/s)
FIG. 1. Simplified cross section of the crustal structure of the Baltic Shield and central Europe based on refraction
seismic P-velocities derived from Figs. 3-4, 3-6, 3-9 and 3-11 in Ansorge et al. (1992) and from the detailed interpretations of FENNOLORA, EUGENO-S (Line 1), EUGEMI and EGT-S86 by Guggisberg et al. (1991), Aichroth et al.
(1992), and Ye (1992).
younger provinces presented in Fig. 1 (lower part). Superimposed on this basically two-layered crust are substantial
lateral variations of the velocity distribution and layer thickness. Details in the uppermost crust of the Scandinavian profile are questionable because of large distances between the
shotpoints in that part of the EGT. The major rocks below the
interface at 20 km average depth and above the M o h o have
been identified as granulites (for correlation of seismic properties with rock species see Kern and Schenk, 1988; Kern,
1990; Rudnick, 1992). The identification of major crustal
rocks in the E U G E M I part of E G T was improved by the investigation of crustal xenoliths in the Tertiary volcanic rocks
(Mengel et al., 1991 ). Luosto et al. ( 1 9 8 9 ) observed that the
lower crust of the Scandinavian Shield is separated into a
felsic upper part and a mafic lower part (the latter with more
than 6.8 k m s i P-wave velocity) at an additional interface.
Mafic granulites mainly from former gabbroic intrusions are
recognized as the most abundant rocks in this lowest crustal
layer of 7 k m average thickness. Confirmation of this identification as mafic granulites comes from areas of the world
where these are sampled by volcanism and investigated as
xenoliths (Rudnick 1992). Rare metapelitic garnet schists
cannot be discriminated from the abundant mafic granulites
by their seismic velocities (Kern, 1990). In the younger fold
belts of Central Europe, mafic granulites occur locally as a
voluminous body but more as a thin layer above the Moho.
Because of difficulties to find out the thickness of the sedimentary top layer of the E G T crust we have used the estimate
by Ronov and Yaroshevskiy ( 1 9 6 8 ) of a 2.9 k m thick unit of
sedimentary rocks in the continental crust. This was based on
drilling and mapping. In the young fold belt of Central Europe, sedimentary basins locally increase to more than 10 km
depth.
The upper continental crust is much better exposed for sampiing and investigation than the lower crust. Two comprehensive chemical studies on the composition of the upper crust
reported comparable results ( S h a w et al., 1967, 1976, 1986;
Taylor and McLennan, 1985). Their results differ distinctly
in Be, B, Cu, Zn, Ba, Ta, and Bi. Shaw et al. (1967, 1976)
analysed 8,512 rocks from large areas of the Canadian Shield
mainly in composite samples. They grouped them into six
lithologic varieties. Taylor and M c L e n n a n ( 1985 ) based their
minor and trace element data on analyses of shales, loesses,
and greywackes in the assumption that the processes of erosion and sedimentation produced an average sample of the
exposed crust. They corrected for element fractionation during weathering and sedimentation from REE distribution and
element ratios, well k n o w n from isotope geochemistry. Their
major element data were adopted from those on the Canadian
Shield composition and their data on Zn, Cd, T1, Pb, and Bi
Ingerson Lecture
1219
Standard Profile of the Continental Crust (EGT)
62 % Archean - Proterozoic Crust (45.5 km thick)
38 % Proterozoic - Phanerozoic Crust (30 km thick)
(Mass of Bulk Cont. Crust: 2.13 x 1019 t)
Sediment. Rocks (14 % U. Cr.)
o2
D
Felsic Intrusives
(10.4 km= 50 % Upper Cr.)
Gabbros
(1.3 km = 6 % Upper Cr.)
/
Sedimentary Rocks* - -
44.0 %
20.9 %
20.3 %
14.6 %
Felsic Intrusives **
50
40
10
Gabbros
(LE MAITRE, 1976)
¢,J
Shales, Siltstones (RONOV and YAROSHEVSKIY, 1968)
Sandstones, Greywackes ( R and Y, 1968)
Mafic etc. Volcanics ( R and Y, 1968)
Carbonates (0.8 % Evaporites) ( R and Y, 1968) ....
% Granites (WHALEN et al., 1987; LE MAITRE, 1976)
% Granodiorites (LE MAITRE, 1976)
% Tonalites (this publication)
(q
cO
Gneisses, Mica Schists
Amphibolites, Marbles
(6.3 km = 30 % Upper Cr.)
Metamorphic Rocks ***
64 %
(Low- and Medium
......._-~ 15.4 %
Grade)
"~
17.8 %
2.6 %
Felsic Granulites
(61.5 % of Lower Crust)
02
Mafic Granulites
(38.5 % o1 Lower Crust)
Gneisses (POLDERVAART, 1955)
Mica Schist (POLDERVAART, 1955)
Amphibolites (POLDERVAART, 1955)
Marbles (RONOV and YAROSHEVSKIY, 1968)
Felsic Granulites
- -
(50 % Archean, 50 % Post-Archean)
(RUDNICK and PRESPER, 1990)
Mafic Granulites
- -
(RUDNICKand PRESPER, 1990)
i<
Abundances in 73 % Geosynclines and 27 % Platforms according to RONOV and
YAROSHEVSKIY (1968)
Abundances according to DALY (1933) and MOORE (1959)
Abundances according to RONOV and YAROSHEVSKIY (1968)
Abundance of Carbonates reduced from 17.6 to 14.6 % according to Carbon Isotope
Balance
FIG. 2. Standard profile of the continental crust derived from the 3000 km long European Geotraverse generalized
on the basis of worldwide mapping, petrological studies and chemical balances (Table 5 etc.).
Table 1 Element concentrations (in ppm) in the Upper Continental Crust (UC) and the Lower
Continental Crust (LC) (For references see footnotes in Table 2)
UC
LC
UC/LC
UC
LC
UC/LC
UC
LC
UC/LC
Si
303480
271330
1.1
Ce
65.7
53.1
1.2
Yb
1.5
2.5
0.60
A!
Fe
Ca
77440
30890
29450
82120
57060
48600
0.94
0.54
0.61
Ni
Nil
La
18.6
25.9
32.3
99
28.1
26.8
0.19
0.92
1.2
U
Br
Ge
2.5
1.6
1.4
0.93
0.28
(1.4)
2.7
5.7
1.0
Nil
Mg
K
25670
13510
28650
21200
31550
13140
1.2
0.43
2.2
Cu
Co
y
14.3
11.6
20.7
37.4
38
27.2
0.38
0.31
0.76
Be
Mo
Eu
3.1
1.4
0.95
1.7
0.6
1.6
1.8
2.3
0.59
Ti
3117
3240
5010
588
0.62
5.5
Nb
26
22
11.3
13
2.3
1.7
Ta
|
1.5
1.4
0.84
0.14
F
CI
Sr
Zr
665
527
953
668
611
640
316
237
872
929
408
568
429
278
352
165
0.75 S¢
0.57 G a
2.3
Pb
1.2
B
1.4
Th
2.3 P r
0.90 [Sm
1.4
HI"
7
14
17
17
10.3
6.3
4.7
5.8
25.3
17
12.5
5
6.6
7.4
6.0
4.0
0.28
0.82
1.4
3.4
1.6
0.85
0.78
1.5
0.62
1.4
0.50
0.75
0.27
0.31
0.102
0.055
0.99
0.6
0.81
0.26
0.43
0.30
0.101
0.080
0.63
2.3
0.62
2.9
0.63
1.0
1.0
0.69
Cr
V
Rb
Zn
N
35
53
110
52
83
228 0.15
149 0.36
41 2.7
79 0.66
34 [2.4
Gd
2.8
2.9
2.5
5.8
2.0
5.4
4.7
2.1
0.8
1.3
0.52
0.62
1.2
7.3
1.5
0.123
0.083
0.061
0.056
0.037
0.170
0.052
0.021
3.3
0.47
1.2
2.7
C
p
Mn
S
Ha
Li
Dy
Sn
Cs
As
Ho
W
Tb
T|
Lu
Sb
Cd
Ag
Bi
Se
In
Hg
1.8
10
K. H. Wedepohl
220
Table 2 Element'concentrations in the Continental Crust (For major elements as oxides see
Table 3 and for references see footnotes)
O
47.2
%a
Nd
27
ppm a
Mo
1.1 ppm o
Si
28.8
%a
Cu
25
ppm a
Br
1.0 ppm P
7.96 % a
Co
24
ppm a
W
1.0 ppm q
AI
a
Fe
4.32 % a
Y
24
ppm
Ca
3.85 % a
Nb
19
ppm a
18
ppm a
Sc
Na
2.36 % a
Mg
2.20 % a
K
2.14 % a
Ti
4010 ppm
C
b
1990 ppm
P
Mn
S
757 ppm a
a
716 ppm
b
697 ppm
Ba
584 ppm c
F
525 ppm
CI
472 ppm
d
d
r
800
ppb
Ho
800
ppb a
Tb
650
ppb a
16
ppm a
TI
520
ppb s
Ga
15
ppm a
Lu
350
ppba
Pb
14.8 ppm a
Tm
300
ppb g
ppm f
Sb
300
ppb t
Th
8.5 ppm a
Se
120
ppb u
Pr
6.7 ppm g
Cd
100
ppb s
Sm
5.3 ppm a
Bi
85
ppb s
Hf
4.9 ppm a
Ag
70
ppb v
Gd
4.0 ppm a
In
50
ppbw
Dy
3.8 ppm g
Hg
40
ppb x
Te
(5)
ppb y
Li
a
I
B
11
Sr
333 ppm a
Cs
3.4 ppm h
Zr
203 ppm a
Be
2.4 ppm
i
Au
Cr
126 ppm a
Sn
2.3 ppm k
Pd
0.4 ppbaa
V
98 ppm a
Er
2.1 ppm g
Pt
0.4 ppb aa
Re
0.4 ppb bb
2.5 ppb
z
Rb
78 ppm a
Yb
2.o ppm a
Zn
65 ppm a
As
1.7 ppm 1
Ru
0A ppb aa
N
60 ppm e
U
1.7 ppm a
Rh
0.06 ppb aa
m
Ce
60 ppm a
Ge
1.4 ppm
Ni
56 ppm a
Eu
1.3 ppm a
30 ppm a
Ta
1.1 ppm n
La
Os
Ir
0.05 ppb
bb
0.05 ppb aa
UC = Upper Crust, LC = Lower Crust
a
UC: SHAW et al. (1967, 1976), LC: RUDNICK and PRESPER (1990) in the proportions of Fig 2
b
Fig. 7
c
UC: calculated from rock averages compiled by PUCHELT (1972) in the proportions of Fig. 2, LC: RUDNIK and
PRESPER (1990)
d
UC: calculated from rock averages complied by WEDEPOHL (1987) in the proportions of Fig. 2, LC: averages of
e
UC, LC: calculated from rock averages compiled by WLOTZKA (1972) in the proportions of Fig. 2 (additional data on
granulites and gabbro
granites HALL, 1988)
f
UC: calculated from rock averages compiled by HARDER (1974) and SHAW et al. (1986) in the proportions of
Fig. 2, LC: data from TRUSCOTT et al. (1986), LEEMAN et al. (1992) and HARDER (1974)
g
interpolated from smooth curve of chondrite normalized REE distribution in the continental crust
h
UC: Rb/19 acccording to MCDONOUGH et al. (1992), LC: RUDNICK and PRESPER (1990)
i
UC: calculated from rock averages compiled by HOERMANN (1969) in the proportions of Fig. 2,
LC: SIGHINOLFI(1973)
k
UC: calculated from rock averages compiled by HAMAGUCH~and KURDODA (1969) and SMITH and BURTON
(1972) in the proportions of Fig 2, LC: RUDNICK anddd PRESPER (1990)
Ingerson Lecture
1221
Table 2. (Continued)
1
UC: calculated from rock averages of ONISH/ and SANDELL (1955), BURWASH and CULBERT (1979) in the
proportions of Fig., 2, LC: gabbro, gneiss minus 20% granite
m UC: calculated from rock averages compiled by HOERMANN (1970) and ONISHI (1956) in the proportions of Fig. 2,
LC: gneiss, gabbro
n
UC: calculated from Nb/Ta = 17.5 (from international reference rocks) GLADNEY et al. (1983), LC: RUDNICK and
PRESPER (1990)
o
UC: calculated from rock averages compiled by MANHE1M and LANDERGREN (1978) in the proportions of Fig. 2,
p
UC, LC: calculated from rock averages compiled by FUGE (1974a) (partly corrected with C1/Br ratio 1000) in the
q
UC, LC: calculated from rock averages compiled by KRAUSKOPF (1970) in the proportions of Fig. 2
gabbro, gneiss minus 20% granite
proportions of Fig. 2. Sedimentary rocks calculated with C1/Br = 290 (seawater)
r
UC: calculated from rock averages compiled by FUGE (1974b) and BECKER et al. (1972) in the proportions of Fig. 2
considering accumulation in C,,~ rich sediments (C/I correlation of PRICE et al. 1970), LC: estimated
s
t
UC, LC: calculated from rock averages ofHEINRICHS et al. (1980) in the proportions of Fig. 2
UC: calculated from rock averages of ONISI-II and SANDELL (1955) and BURWASH and CULBERT (1979) in the
proportions of Fig. 2, LC: estimated
u
UC, LC: calculated from rock averages of KOLJONEN (1973) and KELTSCH (1983) in the proportions of Fig. 2,
S/Se in crustal rocks except sediments: 8.5 x 103
v
UC, LC: calculated from rock averages of HAMAGUCHI and KURODA, (1959) in the proportions of Fig. 2, Cu/Ag
in felsic rocks 300 to 450
w
UC, LC: calculated from rock averages compiled by LINN and SCHMITT (1972) and contributed by VOLAND (1969)
in the proportions of Fig. 2
x
UC, LC: calculated from rock averages compiled by the present author partly from MAROWSKI and WEDEPOHL
y
UC, LC: order of magnitude estimated from data on international reference rocks (GLADNEY et al., 1983,
z
UC: calculated from rock averages compiled by CROCKET (1974) in the proportions of Fig. 2, LC partly from
(1971) and partly from unpublished data of HEINRICHS in the proportions of Fig 2
SIGHINOLFI et al. 1979) and on MORB (HERTOGEN et al. 1990)
SIGHINOLFI and SANTOS (1976) partly from gabbro
aa
UC, LC: from composite sample of 17 European greywackes analysed by ICP MASS in nickel sulfide extract after Te
coprecipitation (HARTMANN, 1995)
bb
UC: ESSER and TLrREKIAN (1993)
f r o m Heinrichs et al. ( 1 9 8 0 ) . T h e y calculated u p p e r crust averages o f Li, Be, B, Ge, As, Se, Zr, M o , Ag, In, Sb, Ba, W ,
and Re f r o m the r e s p e c t i v e c o m p i l a t i o n s edited by W e d e p o h l
( 1 9 6 9 - 1 9 7 8 ) . T h e f o r m e r authors obtained a granodioritic
bulk c o m p o s i t i o n for the u p p e r continental crust. Large sets
o f s e d i m e n t a r y rock s a m p l e s f r o m g e o s y n c l i n e s and platforms
have b e e n a n a l y s e d by R o n o v and Y a r o s h e v s k i y ( 1 9 6 8 ) to
get an a v e r a g e c o m p o s i t i o n o f the s e d i m e n t a r y top o f the
crust. This has e s p e c i a l l y a c c u m u l a t e d s o m e rather m o b i l e
c o m p o u n d s , partly o f volatile e l e m e n t s ( C a C O 3 , NaC1, e t c ) .
A c o m p r e h e n s i v e c o m p i l a t i o n o f c h e m i c a l data on granulites
as the m o s t a b u n d a n t l o w e r crustal rocks has b e e n p u b l i s h e d
b y R u d n i c k and P r e s p e r ( 1 9 9 0 ) .
A distinct d i s c r e p a n c y in estimates about the size and c o m position and in a s s u m p t i o n s about the g e n e s i s o f the l o w e r
and the bulk crust exists b e t w e e n the widely used report o f
T a y l o r and M c L e n n a n ( 1 9 8 5 ) and the results o f our p r e s e n t
investigation. T h e s e authors a s s u m e that the l o w e r crust c o m prises 75% o f the total crust in contrast to our a b o v e listed
50% estimate b a s e d o n the E G T s e i s m i c profile t h r o u g h Europe. T a y l o r and M c L e n n a n ( 1 9 8 5 ) calculated the average
l o w e r crust c o m p o s i t i o n by r e m o v a l o f their u p p e r from their
bulk crust c o m p o s i t i o n . T h e y got a l o w e r crust w h i c h is distinctly m o r e mafic ( 5 4 . 4 % SIO2) than average diorite or andesite ( 5 7 . 7 % SIO2; Le Maitre, 1976). T h e f o r m e r authors
m o d e l l e d a dioritic continental crust f r o m their A r c h e a n crust
c o m p o s i t i o n ( 7 5 % ) and f r o m island arc andesites ( 2 5 % ) .
Their A r c h e a n crust consists o f a 2:1 mixture o f mafic
( M O R B ) and felsic (tonalitic to t r o n d h j e m i t i c ) e n d m e m b e r s ,
w h i c h is in R E E c o m p o s i t i o n c o n f o r m a b l e with andesites ( o r
d i o r i t e s ) . The m o d e l p r o p o s e d by Taylor and M c L e n n a n
( 1985 ) o b v i o u s l y o v e r e s t i m a t e s the a b u n d a n c e s o f andesites
and diorites at active continental m a r g i n s and in the bulk crust.
T h e s e authors stated that their m o d e l is constrained by the
a s s u m p t i o n o f about 4 0 % o f the total heat flow at the average
E a r t h ' s surface b e i n g d e r i v e d f r o m the crust and 60% f r o m
the mantle. T h e y have calculated that for the r e s p e c t i v e heat
p r o d u c t i o n a bulk crust a b u n d a n c e o f 9000 p p m K, 3.4 p p m
Th and 0.9 p p m U is required. I n f o r m a t i o n f r o m refraction
seismic profiles on the crustal structure and c o m p o s i t i o n is
e x p e c t e d to be m o r e reliable than crustal m o d e l s b a s e d on
heat p r o d u c t i o n o f the silicate earth.
1222
K.H. Wedepohl
Tonalltes/Continental
Greywaekes/Continental
Crust
Crust
10.0--
--10.0
8.0--
8.0
6.0--
--
6.0
4.0 B
--
4.0
3.0
--
3.0
--
2.0
m
2.0m
AM
P
Sr,Ga
1.0
4
0.8__
Na Pr,Tb,Yb
Ti,A ,Eu,Gd,Lu
Si,Fe,Ba,V,Li~Pb,Sn,U,Ta
K,Rb,Zr,La,Cu,Th
Zr
Mn,Ca,Zn,Ce,Nd,Y,Sm,H f,Dy,Er,Cs
-
Si,Zn,La,Y,Ga,U
Ti,Mn,V,Ce,Cu Sc,Pb,Th,Gd,Er,Yb,Ho,Tb,Lu
K
P,Ba,Cr,(Hf),Cs
Mg,Co,Sc
Mg,Sr,Co
0.6 m
Nb
/
m
1.0
--
0.8
--
0.6
--
0.4
--
0.3
Ca
Ni,Nb -
0.4 D
0.3
__
AI,Fe,Na,Nd,Pr,Sm,Dy,Eu,Rb
Cr, Ni
0.2--
--
0.1 m
--
0.2
0.1
FIG. 3. Average composition of tonalites and greywackes from worldwide sampling (Table 3) in comparison with
the composition of the continental crust.
CRUSTAL CONCENTRATIONS OF MAJOR ELEMENTS
AND COMMONLY ANALYSED MINOR ELEMENTS
We have recalculated the composition of the continental
crust in the proportions of upper crust to lower felsic crust to
lower mafic crust being equal to about 1:0.6:0.4 as derived
from the EGT refraction seismic profile through Europe. The
precise proportion of these units and of additional subunits in
a standard crustal profile is listed in Fig. 2. The abundance of
major sedimentary, magmatic, and metamorphic rocks in the
subunits of the continental crust is derived from the evaluation
of mapping and drilling reports (mainly from Ronov and Yaroshevskiy, 1968; Daly, 1933; Moore, 1959; supplemented by
additional authors). The chemical composition of the upper
and the lower continental crust is for the majority of commonly analysed elements taken from Shaw et al. ( 1976, 1986)
and from Rudnick and Presper (1990), respectively. The latter authors presented averages of Archean and post-Archean
felsic granulites from terrains ( which we used in a proportion
of I : 1 ) and of mafic granulites sampled as xenoliths. Our data
for the upper and lower continental crust and for the bulk
continental crust are presented in Tables 1 and 2, respectively.
These tables also contain data on additional elements which
were not considered in the above-mentioned references. They
will be handled in the next section. Our bulk crust has a tonalitic composition (cf. Fig. 3) and is also close to the average
of felsic granulites of Rudnick and Presper (1990) and not
much different (except Fe and Ta) from the felsic granulites
of Shaw et al. (1986). The continental crust calculated by us
(Tables 1 and 2) has about twice the concentration of heat
producing elements of the crustal model suggested by Taylor
and McLennan ( 1985 ).
There exists ample experimental evidence (Rushmer, 1991 ;
Wolf and Wyllie, 1991; Springer and Seck, 1995; Rapp,
1994) that tonalites are partial melting products of mafic rocks
under water-undersaturated conditions. This suggests that a
part of the felsic lower crust has formed from tonalitic partial
melts derived from the mafic lower crust after supply of water
from mafic intrusions and dehydration of subducted ocean
crust. This water might have introduced a certain fraction of
K and other incompatible elements into the volume of partial
melting. A part of the felsic lower crust is residual from the
formation of granitic partial melts during orogenic heating.
We have calculated the composition of a felsic lower crust
fragment after subtraction of 30% granitic partial melt (1:1
mixture of I- and S-type felsic granites of Whalen et al.,
1987). The residuum has a major element composition close
to diorites as listed in Table 4. The minor elements Sc, Cr,
Co, Ni, and HREEs increased and Rb, Cs, Pb, Th, and U
decreased in the residuum relative to the felsic lower crust
after granitic melt separation. The minor dioritic residuum
stays in the predominantly tonalitic substratum of the felsic
lower crust, which preserves its overall tonalitic composition.
Greywackes approach on average the chemical composition of tonalites with the exception of B, Zr, Si, Ca, P, and Cr
(Table 3, Fig. 3). This characterizes them as major products
of erosion at active continental margins and indicates the importance of tonalites as early products of magmatism generating new continental crust at plate boundaries. The average
metamorphic rocks of the continental upper crust (64%
gneiss, 15% mica schist, 18% amphibolite; cf. Fig. 2) and
especially their major proportion of gneisses are close in
chemical composition to tonalites and greywackes, the latter
Ingerson Lecture
with the exception of SiO2. The chemical connection indicates
genetic relationship.
A critical proof of the representativeness of the Canadian
Shield major element data for the composition of the upper
continental crust is presented in Table 5. In this table chemical
compositions of major subunits of the upper continental crust
are summed up with the statistical weight of their abundance
as listed in Fig. 2. The sum of the subunits in Column H is
compared with the Canadian Shield data of Shaw et al.
(1976). There exists a good correspondence between the two
sets of data with a little more than 10% deviation in SIO2,
TiO2, Fe203, and Na20. The deviation in Na20 decreases but
does not disappear if the seawater reservoir of this element is
considered. The deviation in SiO2 might be caused by the
mobility of this compound in hot water and its accumulation
in quartz veins. Some other minor deviations could depend
on our choice of data for rock units in Table 5. The good
correspondence has encouraged us to use the Canadian Shield
as representative of the upper continental crust. It additionally
suggests to employ the rock abundances of Fig. 2 as statistical
weight for the calculation of minor element abundances not
considered in the investigations by Shaw et al. ( 1967, 1976,
1986) in a later section.
After a critical proof of the major set of data on the upper
crust we have to test the reliability of our assumptions on the
bulk crust. For this test we use the experience about the crustal
and bulk earth distribution of several radioactive and radiogenic isotopes. Some elements which have important radioactive/radiogenic isotopes are connected by close crystal
chemical relations. Rb, Th, U, K, and Pb are highly incompatible in mantle minerals and therefore extremely accumulated in the continental crust relative to the primitive mantle
(factor hundred; cf. Fig. 5). Pairs of element abundances in
the continental crust, which are neighbors or almost neighbors
in the sequence of incompatibility of Fig. 5 form element
ratios in which they do not differ much from the respective
primitive mantle ( P M ) ratios, the latter as reported by Hofmann (1988). The T h / U ratio is 5 in the calculated crust and
4 in the PM. The more incompatible Th being more enriched
in the crust. Asmerom and Jacobsen (1993) have calculated
an average T h / U ratio of 4.1 from the lead isotopic composition of suspended river loads with a Nd model age from
0.5-3.5 Ga. This result although conformable with our T h / U
ratio of the upper crust (Table 1 ) might suggest a slightly
smaller ratio than 5 for the bulk crust. Our K / U ratio is 1.26
× 104 for the bulk crust compared with 1.27 × 104 in the
PM. The U / P b and K / P b ratios are 0.115 and 1445 in the
crust and 0.116 and 1475 in the PM, respectively. The K / R b
ratio is 274 in the crust and 482 in the PM with the less mantle
compatible Rb being more enriched in the crust. The R b / S r
and U / P b systems important for the crust are interconnected
by the K / R b and K / U relations. The crustal behavior of Rb,
Pb, T1, and several other elements is controlled by their crystal
chemical connection to K and their affinity to the feldspar and
mica structures. Minor differences in this behavior between
Rb, Pb, and T1 are reflected in the different slopes of element
distributions in major rock types plotted in Fig. 6.
The link between the R b / S r and S m / N d systems is the
comparable mantle incompatibility of Sr and Nd (Fig. 5 ). The
1223
S r / N d ratios for the continental crust and the PM are 12.3 and
15.3 with the slightly more compatible Sr causing the lower
crustal ratio. We have used the element or isotope ratios of
worldwide river loads to check our calculated bulk crust (Table 2) and upper crust (Table 1 ) element ratios (Fig. 4). Erosion has apparently favoured the upper crust slightly more
than the lower crust as reflected in the R b / S r and U / P b systems. In a plot of chondrite normalized REE concentrations
the elements with high analytical precision give a smooth
curve. We have corrected the REEs with lower analytical precision and have estimated the Pr, Dy, Er, and Tm concentrations, which were not analysed in the Canadian Shield, by
interpolation in that curve. The final values are presented in
Table 2. Our crustal L u / H f ratio is the same as that reported
by Patchett et al. (1984) for common sedimentary rocks (except pelagic deposits) and used for their isotope studies.
The check by isotope constraints suggests that the use of
the data by Shaw et al. ( 1967, 1976) on the upper crust and
those by Rudnick and Presper (1990) on the lower crust in
the proportions derived from the EGT profile is recommended
for a calculation of the bulk crust composition.
CRUSTAL CONCENTRATIONS OF RARELY ANALYSED
ELEMENTS
Table 2 contains major element and thirty-two minor element data from our investigation in the former section. The
gaps in Table 2 to be filled in this section concern thirty-three
additional minor elements. For two of them (Ba, Ta) Shaw
et al. ( 1967, 1976) and Taylor and McLennan (1985) report
very different concentrations ( 1070 and 550 ppm Ba, 5.7 and
2.2 ppm Ta, respectively) for the upper crust. The average Ba
concentrations analysed in the Canadian Shield are even different between Shaw et al. ( 1967, 1976) and Eade and Fahrig
(1971) (1070 and 810 ppm Ba). If we use the rock abundances listed in Table 2 and Ba averages for common rock
species as reported by Puchelt (1972) we get an upper crust
value of 668 ppm Ba which is intermediate between the upper
crust value of Taylor and McLennan ( 1985 ) and the Canadian
Shield concentration of Eade and Fahrig (1971). Our B a / U
(344) and B a / T h (68.7) ratios are slightly higher and lower
than the respective PM ratios (298, 74.4; Hofmann, 1988),
which indicates that either the mantle incompatibility of Ba
has to be positioned (in deviation from Fig. 5) between Th
and U or that the crustal abundance of Ba could be slightly
higher. In every case should the crustal B a / R b ratio be lower
than the PM ratio of 11.3. It could be a little higher than our
present value of 7.5. If we calculate the Ta concentration in
the upper crust from the Nb concentration by the use of the
PM ratio of 17.6 (Hofmann, 1988) the value of 1.5 is still
slightly lower than 2.2 ppm Ta reported by Taylor and
McLennan (1985). Our calculation is supported by the fact
that analytically very precise data on seven standard reference
rocks (granites, basalts, andesite) compiled by Gladney et al.
(1983) have on average a N b / T a ratio of 17.7. Older data on
common rocks as those compiled by Wedepohl (1978) give
lower N b / T a ratios closer to 10 than to 17.6. This is probably
an analytical problem.
For two of the rather abundant minor elements (C and S),
for which we have not many reliable data on magmatic and
224
K. H. Wedepohl
Table 3 Average concentration of elements in tonalites and greywackes from worldwide
sampling in comparison with the Continental Crust
SiO2
Ti02
A1203
Fe203
MnO
A
Continental
crust
B
Tonalites
61.5%
0.68
15.1
6.28
61.9 %
69.1%
0.77
0.72
C
Greywackes
B/A
1.00
1.13
CIA
1.12
1.06
16.3
13.5
1.07
0.89
6.3
5.9
1.00
0.94
o. 10
0.09
0.10
0.90
1.00
MgO
3.7
2.6
2.3
0.70
0.62
CaO
5.5
4.9
2.6
0.89
0.47
Na20
3.2
3.9
3.0
1.22
0.94
K20
P205
Ua
Sr
Zr
2.4
2.O
0.79
0.83
11.l 8
584 ppm
333
203
0.13
/I.72
Cr
0.70
1.9
0.26
608 ppm
439
173
302
1.44
1.04
1.32
0.85
126
38
88
(I.30
V
98
1113
98
1.(16
1.00
Rb
78
64
72
0.82
0.92
Zn
65
61
76
0.94
1.16
Ce
60
53
58
0.89
0.96
Ni
56
19
24
o.34
0.43
La
30
23
34
0.77
1.13
Nd
27
25
25
0.93
0.93
Cu
25
19
24
0.76
0.98
Co
24
16
15
(1.67
0.63
Y
Nb
24
22
26
0.93
1.08
I).46
0.44
Li
18
19
Sc
11
16
0.67
Ga
Pb
B
16
15
19
16
1.30
1.08
14.8
t4.2
14.2
0.95
0.95
Th
8.5
6.4
9.0
0.76
1.06
Pr
Sm
Hf
Gd
67
8.5
6.1
0.91
5.3
4.9
4.6
4.9
4.6
3.5
4.2
4.0
Dy
4.o
3.8
3.5
3.4
1.27
0.92
0.93
1.05
i0.92
Cs
3.4
3.2
2.2
!/}.93
0.65
Sn
2.3
2.4
Er
2.1
1.9
2.2
Yb
2.0
1.7
2.4
1.7
2.1
2.0
1.3
1.4
1.1
900 ppb
750
370
1.2
U
En
Ta
He
Tb
Lu
An
Pd
Pt
Ru
19
8.8
1.1
650
350
2.5
0.4
0.4
0.1
2111
8.4
0.73
0.60
1.49
1.06
37
11
800 ppb
426 ppm
780 ppb
630
370
4.8
0.4
0.4
0.1
1.011
3.3
1.04
0.90
1.20
1.00
1.06
1.00
I. 11
1.15
1.06
0.86
0.71
1.00
0.911
1.02
1.05
1.16
0.88
0.98
0.96
1.06
Ingerson Lecture
1225
Table 3 (continued)
A
Continental
B
Tonalites
C
Greywackes
B/A
( 7A
crust
Rh
0.06
0.06
Ir
0.05
0.05
Tonalites: Total average of group averages from USA, Canada, Sri Lanka, Greenland, Finland, UK, Portugal, with equal
statistical weight: ARTH et aL (1978), ERMANOVICS et al. (1979), TARNEY et al. (1979), SCHERMERHORN (1987
and person, comm.), PARADIS et al. (1988), POHL and EMMERMANN (1991), WEDEPOHL et al. (1991), TEPPER et
al. (1993)
Gre'~caekes: Total average of group averages from USA, Canada, Australia, India, Sri Lanka, Germany with equal
statistical weight: WEBER (1960), ONDRICK and GRIFFITHS (1969), TAYLOR and McLENNAN (1985), POHL and
EMMERMANN (1991), WEDEPOHL (composite 17 greywackes, Germany, unpubl.)
metamorphic rocks, we have provided a stable isotope balance
to improve the bulk crust abundance values (Fig. 7 ) . These
elements occur mainly in two valence states (with very different isotopic characteristics) and are highly accumulated in
the sedimentary top of the crust. The bulk crust isotope data
of the two balances must be close to the most probable values
for the mantle. The carbon balances suggest that the proportion of carbonate rocks in the sedimentary column as calculated by Ronov and Yaroshevskiy ( 1 9 6 8 ) , has to be reduced
from 17.6 to 14.6%. The sulfur balance suggests a value for
the abundance o f SOn in evaporites. This value is very important for an evaluation of the crustal distribution of C1. The
high mobility of NaC1, its abundance in c o m m o n sedimentary
rocks and the scarcity of data on chlorine in magmatic and
metamorphic rocks are some o f the difficulties for the calculation of a reliable value of the CI abundance in the continental
crust which we have investigated in detail (Wedepohl, 1987).
This element is highly accumulated in the sedimentary top of
the crust ( 5 5 % ) whereas F occurs mainly in the magmatic
and metamorphic part of the crust where it replaces OH in
mineral structures. The other elements which are especially
accumulated in the sedimentary cover are I, Br, B, Hg, N, and
As which form mobile a n d / o r volatile compounds (Fig. 8).
The crustal abundances in Table 2, which were not reported
in the former section, are all calculated from literature data
(foot-notes in Table 2) on c o m m o n rocks with the statistical
weight of the rock abundances in the upper and lower crust
respectively (Fig. 2). For some of the elements (Be, Ge, Mo,
I, Sb, In, Te, Pt-group, and Re) reliable literature data on
c o m m o n rocks, especially on metamorphic rocks, are rather
rare. Sample contamination is a serious problem with several
of the rare elements (I, Hg, Pt-group, etc.). Knowledge about
the crystal chemical behavior of some rare elements allows a
separate check on their crustal abundance. In c o m m o n rocks
the following element ratios are almost constant: S i / G e and
S i / S n - 105, S / S e ~ 1 0 4, A s / S b ~ 6, C I / B r ~ 300 (sedi-
Table 4 Balance of residual rocks after 30% partial melting
of felsic lower crust
A
Felsic lower
crust
[63.1
SiO~
0.70
TiO~
AI:O3
Fe:03
MnO
MgO
B
Granites
fl and S
felsic)
73.4
0.27
C
Felsic lower
crust minus
30% B
58.7
0.88
D
Diorites
57.5
0.95
14.86
13.5
15.4
16.7
6.89
2.0
8.9
8.0
0.10
0.04
0.12
0.12
3.60
0.57
4.9
3.7
CaO
4.83
1.50
6.3
6.6
Na20
3.17
3.10
3.9
3.5
K~O
2.11
4.35
1.2
1.8
P20s
0.18
0.10
0.21
0.29
Reference:
RUDNICK
and
PRESPER(1990)
WHALEN ~
a1.(1987)
LEMAITRE
(1976)
1226
K . H . Wedepohl
Table 5 Balance of major rock species in the Upper Continental Crust
SiO:
TiO2
AI:O3
Fe203
MnO
MgO
CaO
Na20
K20
P20s
H~0 +
CO:~
Reference
A
B
C
Sedimentary
rocks
(propoa.acc.
Fig. 2)
Granites
(I and S
felsi¢)
Granodiorites
52.4
0.68
13.7
6.4
0.12
3.3
9.0"
1.65
2.14
0.18
73.4
0.27
13.5
2.0
0.04
0.57
1.50
3. I0
4.35
0.10
0.6
66.1
0.54
15.7
4.4
0.08
1.74
3.83
3.75
2.73
0.18
0.8
3.1
D
61.9
(/.77
16.3
6.3
0.10
2.6
4.9
3.9
1.9
0.26
0.7
5.9*
RONOVand
WHALEN et LEMAITRE
YAROSHEVSKIY al. (1987)
(1976)
(1968) *corrected
acc. Fig 7
Sm
Nd
Lu
Hf
Sm/Nd
147Sm / 144Nd
Lu/Hf
176Lu / 177Hf
G
Gabbros
Metamorphic
Rocks (proport. acc.
Fig. 2)
Upper continental 15 % A, 25% B, 20%
crust
C, 4% D,
(Canad. Shield)
6% E, 30% F
50.1
1.10
15.5
11.5
0.12
7.6
4.58
2.4
0.9
0.24
0.7
63.7
0.74
15.1
5.96
0.10
2.91
3.88
2.87
3.21
0.22
0.5
1.1
POLDERVAART
(1955)
64.9
0.52
14.6
4.4
0.07
2.2
4.1
3.5
3.1
0.15
0.8
LEMAITRE
(1976)
m e n t s e x c l u s i v e l y ) , K / T I -- 4 x l 0 4 (cf. Fig. 6 ) , K / B i - 2.5
× l 0 s, R b / C s = 19 ( e x c e p t feisic l o w e r crust; M c D o n o u g h
et al., 1 9 9 2 ) , C u / A g ~ 4 x 10 2 ( e x c e p t m a f i c r o c k s ) , C u /
A u ~ 6 x 10 3 ( e x c e p t m a f i c r o c k s ) . T h e s e q u e n c e o f a b u n d a n c e s o f P t - g r o u p e l e m e n t s in p r i m i t i v e u l t r a m a f i c r o c k s is
c o n t r o l l e d b y their c o s m i c a b u n d a n c e a n d the o c c u r r e n c e o f
sulfides: Pt > R u > P d > Ir ~ O s > R h . U n d e r the r e d o x
c o n d i t i o n s in the c r u s t b e i n g d i f f e r e n t f r o m t h o s e o f the P M
the s e q u e n c e c h a n g e s to Pd -- Pt > R u > R h ~> O s , Ir. T h e
e l e m e n t M o b e i n g a l m o s t as c o m p a t i b l e as Pr is e x p e c t e d to
h a v e a c r u s t a l M o / P r ratio c l o s e to that o f the P M (0.23; N e w s o m et al., 1 9 8 6 ) . O u r ratio o f 0 . 1 6 f r o m T a b l e 2 i n d i c a t e s
that M o is s l i g h t l y less i n c o m p a t i b l e at partial m e l t i n g in t h e
K / Rb
= 255
Rb
78 ppm
Sr
333 ppm
Rb/Sr
= 0234 (0 348 UC, 0.116 LC)
87Rb/86Sr=067
E
SHAW el al.
(1967, 1976)
64.0
0.59
14.7
4.9
0.08
2.4
14.1
2.9/3.1"*
3.1
0.17
1.0
0.80
Fig. 2
** includingoceans
m a n t l e t h a n Pr. T h e correct o r d e r o f m a g n i t u d e c o n f i r m s the
base of our calculated Mo abundance.
GENETIC IMPLICATIONS
T h e e x p l a n a t i o n o f t h e f o r m a t i o n o f the c o n t i n e n t a l c r u s t
h a s to p r o c e e d f r o m the fact that it c o n s i s t s m a i n l y o f tonalites
a n d to a m i n o r d e g r e e o f granites, g r a n o d i o r i t e s , a n d m a f i c
rocks. T h e p r i m a r y m a g m a s for its g e n e s i s w e r e m a f i c partial
melts from the upper mantle. The concentrations of highly
i n c o m p a t i b l e e l e m e n t s in the c r u s t e x c e e d t h o s e o f the p r i m itive m a n t l e b y a factor o f h u n d r e d (Fig. 5 ) . M O R B m a g m a s
as the m o s t a b u n d a n t partial m e l t s f r o m t h e p r e s e n t u p p e r
5.3 ppm
27 ppm
0.35 ppm
4.9 ppm
= 0196 (018UC. 021 LC)
= 0.119
= 0071 (0046UC, 0 1 0 7 L C )
=0.10
K/U
= 1.2x 104
U
1.7 ppm
Th
8.5 ppm
Pb
14.8 ppm
Th / U
= 5.0" (41 UC, 7.1 LC)
238U / 204pb ** = t4 = 8.4
(~ Upper Crust 10.9; It Lower Crust 5.3)
*An Upper Crust value of 4.1 is suggested by crustal Pb
isotope evolution (STACEY and KRAMERS, 1975) and by
load of worldwide rivers (ASMIEROM and JACOBSEN,
1993)
87Rb/86Sr = 0.81
87Sr / 86Sr = 0.716
in load of worldwide rivers draining areas
with average model age of 2.1 Ga
(GOLDSTEIN and JACOBSEN, 1988)
**206pb / 204pb + 207pb / 204pb +208pb / 204pb = 73
(ZARTMAN and HAINES, 1988)
Sm / N d = 0.19
in load ol worldwide rivers
(GOLDSTEIN and JACOBSEN, 1988)
Lu / Hf 0.07 in common sedimentary
rocks, e~'cept pelagic clays
(PATCHETT et al, 1984)
i~BSE ~ 8 (BSE = Bulk Silicate Earth)
(WHITE, 1993)
Resemblance ~aCont.Cr, ~ I~BSE
because of high depletion of U and Pb in the Upper Mantle
FIG. 4. Average composition of the continental crust in elements with important radioactive and radiogenic isotopes.
Comparison with worldwide river loads.
Ingerson Lecture
~ I00
o
°°o
100
P°b
.~ 60
~ 40
o
Nb,Ta
Q.
N 20
Oo
6O
40
o
o
Hf Zr
o o
o
2O
o
~a
°o
~ 6.0
2
production of the early upper mantle has probably formed the
primordial felsic cores of the continents (McCulloch, 1993).
At increasing thickness of the continental crust this process
of formation of tonalitic partial melts shifted into the marie
lower crust and was controlled by water from mafic intrusions
and dehydration of subducted ocean crust. A certain thickness
of the tonalitic crust and heat advection has allowed its partial
melting to form granitic magmas. Melt separation and uprise
of felsic magmas from their source was the major process of
chemical fractionation within the continental crust. Additional
fractionation was caused by weathering and selective water
transport of the weathering products to form greywackes,
clays, sands, carbonates, and evaporites. Camir6 et al. ( 1 9 9 3 )
have calculated that Archean greywackes from the Canadian
Shield are derived from about 65% tonalites/trondhjemites,
30% tholeiitic basalts, and 5% komatiites which is almost
identical with the lower crust composition derived by us
(Fig. 2).
The almost smooth curve of primitive mantle normalized
element concentrations of Fig. 5 is based on a certain sequence of increasing compatibility of elements in the structure
of c o m m o n mantle minerals. This was derived by H o f m a n n
( 1 9 8 8 ) and Sun et al. ( 1 9 7 9 ) from element behavior at
M O R B formation. The smooth part of the curve of Fig. 5
reflects partial melting of upper mantle rocks as the major
- 200
m2oo
ooo
4.0
10
o
o
Ti
ooo
o
6.0
4.0
o
2.0-
2.0
o
o
i.o-
1.0
o o
-
0.6•=- 0.4 -
1227
0.6
-- 0.4
Ba U K Ce Pr Sir HI Ti Gd Dy Y T~n L~ A, Sc Fie
li I
I
I
I
I
I
I
I
II I
t
I
I
I
I
Rb Th Nb La Pb Nd S m Zr Eu Tb Ho Er Y b Na Ca Cu
<
Ta
Increasing Incompatibility of Elements in Mantle Rocks (according to HO F MANN, 1988)
~5
FIG. 5. Average composition of the continental crust in selected
elements normalized to the primitive mantle (ace. to Hofmann, 1988).
The elements are plotted in the sequence of increasing compatibility.
mantle are characterized by a large depletion of these highly
incompatible elements which is complimentary to their accumulation in the crust ( H o f m a n n , 1988).
Extensive partial melting in a subducted fraction of the
early marie crust which was caused by the relatively high heat
K/Rb
-2
10 K/Pb
-3
10 K/TI
1000 --[
600
- 1000
-
-
600
Continental Crust
400
-
_
400
200
-
-
200
100
-
MORB
100
Continental Crust
60-
60
40
40
Continental Crust
20_
~--
20
MORB
10-
ppm Rb, Pb, ppb TI
6
I
0.6
I
1
I
2
t ,°
6
I
I
I
4
6
10
I
I
20
40
I
I
60
100
I
200
I
I
400 600
I
1000
I
2000
FIG. 6. Correlation of K/Rb, K/Pb and K/TI ratios with Rb, Pb and TI concentrations in abundant crustal rocks (data
from Heinrichs et al. 1980, MORB data on K, Rb, Pb from Hofmann, 1988).
1228
K.H. Wedepohl
Carbon
ppm C
Sulfur
Mass
Proportion
513 C %0 *
7.4 %
Sediment. R.: Carbonates 16000** ~
79.7%
Reduced Carbon 4000*** J"
26.5 % Felsic Intrusives
360
5.1%
3.2 %
Gabbros
220
0.4%
15.9 % Gneisses, Schists, etc.
820
7.0%
28.9 % Felsic Granulites
820
5.6%
18.1% Mafic Granulites
220
2.2%
0
-25
-15
-13
-13
ppm S ~
Sedimentary Rocks:
(Excluding Evaporites
Including Evaporites
26.5 % Felsic Intrusives
3.2 % Gabbros
15.9 % Gneisses, Schists, etc.
7.4 %
%
%
%
%
-15)
-6.7
+1.4
+0.5
+3.0
953
75.3 %
-2.8
350
500
11.1%
13.6 %
+3.0
+0.5
Lower Continental Crust
408
24.7 %
+1.8
Continental Crust
697
Upper Continental Crust
Continental Crust
1990
(4.2xl 016 t C)
***
....
2860
3810
334
780
690
42.3
13.3
3.5
16.2
-6 %o ....
28.9 % Felsic Granulites
18.1% Mafic Granulites
**
Mass
~7
Proportion 534 S %.
Abundances mainly from HOEFS (1973), SCHWARCZ
(1969), VEIZER and HOEFS (1976), JUNGE et al.
(1975), HOEFS (1969)
Abundance of Sedimentary Carbonates reported by
RONOV and YAROSHEVSKIY (1968) reduced from
22500 to 16000 ppm C to adjust the proportion of
Carbonate to Reduced Carbon to 4 : 1
RONOV (1958)
The Mantle Source is often reported to contain 513 C
from -5 to -6 %o
(1.44x1016 t S) -1.66
7 Abundances mainly from RONOV el al. (1974), RONOV and
YAROSHEVSKIY (1968), KAPLAN et al. (1963), SIEWERS
(1974), SCHNEIDER (1978), WEDEPOHL (1987), WEDEPOHL
et al. (1991)
q~7
Crustal Sulfur must be supplemented by Seawater Sulfur
( 1 . 4 x 1 0 1 5 t S ; 5 3 4 S + 2 0 %o ) to get 834 S + 0.3 %o
(Mantle Source,
not far from
Meteoritic Troilite)
FIG. 7. Balances of C and S and their isotopic composition in the continental earth's crust.
process of incompatible element accumulation in the crust. A
few deviations from this character can be used to identify
additional processes partly connected with the formation of
tonalites. The large accumulation of Rb, Ba, K (and C s ) in
the continental crust indicates that residual K-feldspar, mica
( a n d amphibole) were unimportant at the formation of tonalitic magmas.
W e explain the required shift of Nb, Ta, and Ti in Fig. 5
to positions of larger compatibility in the curve of element
distribution to balance the depletion as due to the existence
of residual rutile at the formation of tonalitic partial melts
(Rapp, 1994). The excess of Zr and H f was caused by the
temperature dependent extraction of these elements by the
tonalitic melts from their mafic source. An average concentration of 200 p p m Z r in the continental crust indicates temperatures of partial melting not far from 850°C. (Watson and
Harrison, 1983). The excess of Pb in the continental crust
was explained by Peucker-Ehrenbrink et al. ( 1 9 9 4 ) as due to
the hydrothermal leaching of the ocean crust and remobilization of leached Pb at the dehydration of subducted ocean
crust into continental magma.
The formation and chemical fractionation of the continental
crust required water. By far the largest amount of crustal water
occurs in the oceans (1.4 x 101st) and the second largest
fraction in the porous volume of sediments (0.3 x 10~st).
The continental crust (excluding the oceans) contains on average about 2% H 2 0 which is mainly accumulated in the upper crust and its sedimentary top. Dehydration of subducted
ocean crust and of hydroxide containing minerals in the continental crust is required for partial melting in the mafic and
felsic lower crust, respectively. Mantle degassing has formed
the atmosphere and the hydrosphere and has controlled weathering at the Earth's surface as well as the accumulation of C,
S, CI, N, Ar, B, As, Br, I, Sb, Se, and Hg.
The compositions of the upper and lower continental crust
are compared in Fig. 8. The elements which form volatile
compounds and the highly incompatible elements characterize
the upper crust and 3-d elements are specifically accumulated
in the lower crust. This fractionation is mainly caused by the
uprise of granitic melts and the preferential intrusion of
tholeiitic basaltic magmas into the lower crust due to their
compressibility (Kushiro, 1982). Both processes are interrelated.
The standard crustal profile of Fig. 2 which was derived
from the European Geotraverse is converted into a cartoon
with symbolized compositions and processes (Fig. 9 ) . Rock
species are drawn almost in their correct proportions with felsic rocks in white and mafic rocks in black. By far the largest
proportion of felsic rocks has a tonalitic composition. It is
formed in the mafic lower crust and intrudes into the upper
together with granitic magmas at orogenic events (Pitcher,
1979). Granitic partial melts from a tonalitic or sedimentary
source have increased in their proportion of the crustal magmatic production during Earth's history. Crustal temperatures
as calculated by C h a p m a n ( 1 9 8 6 ) are listed in the left part of
Fig. 9. In our test area of the EGT, the heat flow density
increases from the Scandinavian Shield ( 4 0 m W m -2)
through Central Europe (60 m W m -2) to the Alpine orogen
(<~90 m W m-2; for details of the M o h o temperature see
Banda and Cloetingh, 1992). Orogenic heating which is re-
Ingerson Lecture
108-
I
Cs
Br
6-
B
Ta, TI
z-
u~
Be, Li
Pb, F, Zr
1.00.8-
Ba, Cd, Sb
Ga, Pr, Sm
"-.
~
0.6-
Zn, Ca,Ti, Mn
Fe
0.4-
0.2-
Bi
U, Rb, Hg
CI, S, Nb, W
Th
Cu
Mg
V
C0
Sc
~
Hf, As
Na, Ce, La, Sn
Si, Ge
AI, Nd Sr
P'Y Ag
Dy, Yb, Eu, Ho,Tb, Lu
Ni
Cr
0.1
I, Br, C, B, Hg, N, CI, S, As accumulated in the Sedimentary Top
of the Upper Crust
FIG. 8. Comparison of the average composition of the upper
with the lower continental crust (Table 1; references in footnotes of
Table 2).
quired for partial melting within the crust was mainly caused
by mafic intrusions. At a heat flow density of 90 m W m 2 the
temperature at M o h o depth is more than 1000°C which allows
water undersaturated partial melting of mafic rocks. The for-
._o
(..)
e-.
• d. ~ 6 .
8'E
=E
°C
°C
1229
mation of water undersaturated granitic melts from abundant
source rocks requires about 800°C ( J o h a n n e s and Holtz,
1990).
In the shield areas with their larger thickness and distinctly
lower temperatures mafic granulites can be converted into
eclogites (Ringwood, 1975). This material with a density
higher than that of lherzolite can sink into the mantle if tectonically separated from the lower crust. This process of delamination, which will exclude very dense upper mantle, is
required to recycle the major part of the mafic lower crust
back into the mantle. The different stages of the mafic lower
crust which was formed during Earth's history must have been
in total much thicker than the present lowest part of the crust
(average 7 km). One single intrusion such as that of the Ivrea
area in northern Italy has about 10 k m thickness. A bulk crust
with about 30 k m tonalite required 90 to 120 k m mafic source
rocks if their degree of partial melting was 25 to 30%. An
open-system behavior of the crust, which was caused by a
major delamination of eclogite with excess Eu 2+, is indicated
by the depletion of Eu 2÷ in the majority of crustal rocks. The
lower crust contains on average a surplus of 0.2 ppm Eu 2÷
(based on data reported by Rudnick and Presper, 1990) and
the upper crust has a deficit of 0.58 p p m Eu 2+ (based on data
compiled by Wedepohl, 1991, and calculated in the rock proportions of Table 5, last c o l u m n ) . This comparison clearly
demonstrates the major depletion of Eu e+ in the bulk continental crust. The result agrees with data from a bulk crust
section in China ( G a o et al., 1992).
"or.
~E
°C
Sedimentary Rocks
km
Diagenesis
150 100
60
5
~= Plutons of Granites
GreenschistFacies
310 200 130
10
C~ Plutons of Tonalites
470 300 180
15
590 350 230
20
700 430 270
25
800 470 300
30
880 550 340
35
970 610 370
40
AmphiboliteFacies
GranuliteFacies
Mafic
Intrusions
t"~
Vp
~
f
~
~ 5.9 - 6.5
(km/s)
~. Mica-Schists, Gneisses,
Amphibolites
(Conrad Discontinuity)
o
.o
Felsic Granulites
Felsic Granulites
I
6.5 - 7.5
(Residual from Anatexis)
(km/s)
Mafic Granulites
(Residual from Anatexis)
Mafic Granulites
Moho Discontinuity
OJ
E
E
E
45
E
o
E
o
E
o
50
Spinel Lherzolites and
Spinel Harzburgites
> 8 (km/s)
Heat Flow Density
FIG. 9. Cartoon of the standard profile of the continental crust with temperature-depth relations of 3 geotherms
calculated by Chapman (1986).
1230
K . H . Wedepohl
T h e crustal c o m p o s i t i o n as calculated in this investigation
is distinctly m o r e felsic than that m o d e l l e d by Taylor and
M c L e n n a n ( 1 9 8 5 ) . T h e a m o u n t o f the m o s t i n c o m p a t i b l e ele m e n t s in the continental crust c a n n o t have b e e n supplied
from the primitive c o m p o s i t i o n o f the upper c o n v e c t i n g m a n tle with 670 k m d e p t h exclusively. M o r e than half o f the
crustal reservoir o f Cs, Rb, and Ba and half o f that o f Th, U,
and K m u s t originate f r o m the l o w e r mantle. T h e separation
o f the c o n v e c t i o n cells o f the u p p e r and l o w e r m a n t l e was
apparently interrupted during certain e p i s o d e s w h i c h requires
future investigations.
Our results on the crustal c o m p o s i t i o n have m a i n l y benefitted f r o m recent research on the l o w e r continental crust.
C o n s t r u c t i v e d i s c u s s i o n s with n u m e r o u s colleagues and assistance in the graphical p r e s e n t a t i o n o f the results were gratefully accepted.
REFERENCES
Aichroth B., Prodehl C., and Thybo H. (1992) Crustal structure along
the Central Segment of the EGT from seismic-refraction studies.
Tectonophysics 207, 43-64.
Ansorge J., Blundell D., and Mueller S. (1992) Europe's lithos p h e r e - s e i s m i c structure. In A Continent Revealed. The European Geotraverse (ed. D. Blundell et al.), pp. 33-69. Cambridge
Univ. Press.
Arth J. G., Baker F., Peterman Z. E., and Friedman I. (1978) Geochemistry of the gabbro-diorite-tonalite-trondhjemite suite of
southwest Finland and its implications for the origin of tonalitic
and trondhjemitic magmas. J. Petrol. 19, 289-316.
Asmerom Y. and Jacobsen S. B. (1993) The Pb isotopic evolution
of the Earth: inferences from river water suspended loads. Earth
Planet. Sci. Lett. 115, 245-256.
Banda E. and Cloetingh S. (1992) Europe's lithosphere-Physical
properties of the lithosphere. In A Continent Revealed. The European Geotraverse (ed. D. Blundell, et al.), pp. 71-80. Cambridge
Univ. Press.
Becker V. J., Bennett J. H., and Manuel O. K. (1972) Iodine and
uranium in sedimentary rocks. Chem. Geol. 9, 133-136.
Burwash R. A. and Culbert R. R. ( 1979 ) Abundance and distribution
of Hg and As in the polymetamorphic Precambrian basement of
western Canada. Canadian J. Earth Sci. 16, 2196-2203.
Camir6 G. E., LaflEche M. R., and Ludden J. B. (1993) Archean
metasedimentary rocks from the northwestern Pontiac subprovince
of the Canadian Shield: Chemical characterization, weathering and
modelling of the source areas. Precambr. Res. 62, 285-305.
Chapman D. S. (1986) Thermal gradients in the continental crust. In
The Nature o f the Lower Continental Crust (ed. J. B. Dawson et
al.), pp. 63-70. Blackwell.
Clarke F. W. (1924) The Data o f Geochemistry. U S Geol. Surv.
Bull. 770.
Crocket J. H. (1974) Handbook of Geochemistry 11, Sections 79,
B-O. Springer-Verlag.
Daly R. A. (1933) Igneous Rocks and the Depth o f the Earth, 2nd
Ed. Hafner.
Eade K. E. and Fahrig W. F. ( 1971 ) Geochemical evolutionary trends
of continental p l a t e s - - a preliminary study of the Canadian Shield.
Geol. Surv. Canada Bull 179, 1-51.
Ermanovics I. F., McRitchie W. D., and Houston W. N. (1979) Petrochemistry and tectonic setting of plutonic rocks of the Superior
Province in Manitoba. In Trondhjemites, Dacites and Related
Rocks (ed. F. Barker), pp. 323-362. Elsevier.
Esser B. K. and Turekian K. K. (1993) The osmium isotopic composition of the continental crust. Geochim. Cosmochim. Acta. 57,
3093 - 31 04.
Fuge R. (1974a) Handbook o f Geochemistry 11, Sections 35 B-G,
I-L, O. Springer-Verlag.
Fuge R. (1974b) Handbook o f Geochemistry 11, Sections 53 B-M,
O. Springer-Verlag.
Gao S. et al. (1992) Chemical composition of the continental crust
in the Qingling Orogenic Belt and its adjacent North China and
Yangtze kratons. Geochim. Cosmochim. Acta 56, 3933-3950.
Gladney E. S., Burns C. E., and Roelandts I. (1983) 1982 compilation of elemental concentrations in eleven United States Geological Survey rock standards. Geostand. Newslett. 7, 3-226.
Goldschmidt V. M. ( 1933 ) Grundlagen der quantitativen Geochemie.
Fortschr. Mineral. Kristall. Petrogr. 17, 112-156.
Goldstein S. J. and Jacobsen S. B. (1988) Nd and Sr isotopic systematics of river water suspended material: Implications for crustal
evolution. Earth Planet. Sci. Lett. 87, 249-265.
Guggisberg B., Kaminski W., and Prodehl C. (1991) Crustal structure of the Fennoscandian Shield: a travel time interpretation of
the long-range FENNOLORA seismic refraction profile. Tectonophysics 195, 105-137.
Hall A. (1988) The distribution of ammonium in granites from
South-West England. J. Geol. Soc. London 145, 37-41.
Hamaguchi H. and Kuroda R. (1959) Silver content of igneous rocks.
Geochim. Cosmochim. Acta 17, 44-52.
Hamaguchi H. and Kuroda R. ( 1969 ) Handbook of Geochemistry 11,
Sections 50 B-M. Springer-Verlag.
Harder H. (1974) Handbook o f Geochemistry 11, Sections 5, B-O.
Springer-Verlag.
Hartmann G. (1995) H~iufigkeit und Verteilung der Platin-Grnppenele-mente und des Goldes im Erdmantel und in der Erdkruste. (in
prep.)
Heinrichs H., Schulz-Dobrick B., and Wedepohl K. H. (1980) Terrestrial geochemistry of Cd, Bi, TI, Pb, Zn, and Rb. Geochim.
Cosmochim. A cta 44, 1519-1523.
Hertogen J., Janssens M. J., and Palme H. (1980) Trace elements in
ocean ridge basalt glasses: implications for fractionations during
mantle evolution and petrogenesis. Geochim. Cosmochim. Acta 44,
2125 2143.
Hoefs J. (1969) Handbook o f Geochemistry I1, Sections 6, E-K, M.
Springer-Verlag.
Hoefs J. ( 1973 ) Ein Beitrag zur Isotopengeochemie des Kohlenstoffs
in magmatischen Gesteinen. Contrib. Mineral. Petrol. 41, 277300.
Hoermann P. K. (1969) Handbook of Geochemistry 11, Sections 4
B-O. Springer-Verlag.
Hoermann P. K. (1970) Handbook o f Geochemistry 11, Sections 32
B-M, O. Springer-Verlag.
Hofmann A. W. (1988) Chemical differentiation of the earth: The
relationship between mantle, continental crust and oceanic crust.
Earth Planet. Sci. Lett. 90, 297-314.
Johannes W. and Holtz F. (1990) Formation and composition of H20
undersaturated granitic melts. In High Temperature Metamorphism and Crustal Anatexis (ed. J. R. Ashworth and M. Brown),
pp. 87-104. Unwin and Hyman.
Junge C. E., Schidlowski M., Eichmann R., and Pietrek H. (1975)
Model calculations for the terrestrial carbon cycle: Carbon isotope
geochemistry and evolution of photosynthetic oxygen. J. Geophys.
Res. 80, 4542-4552.
Kaplan I. R., Emery R. O., and Rittenberg S. C. (1963) The distribution and isotopic abundance of sulphur in recent marine sediments off southern California. Geochim. Cosmochim. Acta 27,
297 331.
Keltsch H. (1983) Die Verteilung des Selens und das Schwefel-Selen-Verhaltnis in den Gesteinen der kontinentalen Erdkruste. Dr.
Dissertation, Univ. Grttingen.
Kern H. (1990) Laboratory seismic measurements: an aid in
the interpretation of seismic field data. Terra Nova 2, 6 1 7 628.
Kern H. and Schenk V. (1988) A model of velocity structure beneath
Calabria, southern Italy, based on laboratory data. Earth Planet.
Sci. Lett. 87, 325-337.
Koljonen T. (1973a) Selenium in certain igneous rocks. Bull. Geol.
Soc. Finland 45, 9-22.
Koljonen T. (1973b) Selenium in certain metamorphic rocks. Bull
Geol. Soc. Finland 45, 107-117.
Krauskopf K. B. (1970) Handbook o f Geochemistry 11, Sections 74,
B-O. Springer-Verlag.
Ingerson Lecture
Kushiro I. (1982) Density of tholeiite and alkali basalt magmas at
high pressures. Annu. Rpt. Director Geophys. Lab. Washington
Yearbook 81, 3 0 5 - 3 0 9 .
L e e m a n W. P., Sisson V. B., and Reid M. R. (1992) Boron geochemistry of the lower crust: evidence from granulite terranes
and deep crustal xenoliths. Geochim. Cosmochim. Acta 56,
775 - 7 8 8 .
Le Maitre R. W. (1976) The chemical variability of some common
igneous rocks. J. Petrol. 17, 5 8 9 - 6 3 7 .
Linn T. A. and Schmitt R. A. (1972) Handbook o f Geochemistry I1,
Sections B-M, O. Springer-Verlag.
Luosto U., Flueh E. R., Lund C. E., and Working Group (1989) The
crustal structure along the polar profile from seismic refraction
investigations. Tectonophysics 162, 5 1 - 8 5 .
Manheim F. T. and Landergren S. (1978) Handbook o f Geochemistry
11, Sections 42, B-O. Springer-Verlag.
Marowsky G. and Wedepohl K. H. (1971) General trends in the
behavior of Cd, Hg, TI and Bi in some major rock forming processes. Geochim. Cosmochim. Acta 35, 1255-1267.
McCulloch M. T. (1993) The role of subducted slabs in an evolving
earth. Earth Planet. Sci. Lett. 115, 8 9 - 1 0 0 .
McDonough W. F., Sun S.-S., Ringwood A. E., Jagoutz E., and Hofmann A. W. (1992) Potassium, rubidium and cesium in the Earth
and Moon and the evolution of the mantle of the Earth. Geochim.
Cosmochim. Acta 56, 1001-1012.
Mengel K., Sachs P. M., Stosch H. G., Woeruer G., and Look G.
(1991) Crustal xenoliths from Cenozoic volcanic fields of West
Germany: Implications for structure and composition of the continental crust. Tectonophysics 195, 271 - 289.
Moore J. G. (1959) The quartz diorite boundary line in the western
United States. J. Geol. 67, 198-210.
Newsom H. E., White W. M., Jochum K. P., and Hofmann A. W.
(1986) Siderophile and chalcophile element abundances in oceanic basalts, Pb isotope evolution and growth of the Earth-s core.
Earth Planet. Sci. Lett. 80, 299-313.
Ondrick C. W. and Griffiths J. C. (1969) Frequency distribution of
elements in Rensseler graywacke, Troy, New York. G.S.A. Bull.
80, 5 0 9 - 5 1 8 .
Onishi H. (1956) The geochemistry of germanium. Bull Chem. Soc.
Japan 29, 6 8 6 - 6 9 4 .
Onishi H. and Sandell E. B. (1955) Notes on the geochemistry of
antimony. Geochim. Cosmochim. Acta 8, 213-221.
Paradis S., Ludden J., and Gelinas L. ( 1988 ) Evidence for contrasting
compositional spectra in comagmatic intrusive and extrusive rocks
of the Late Archean Blake River Group, Abitibi, Quebec. Canadian J. Earth Sci. 25, 134-144.
Patchett P. J., White W. M., Feldmann H., Kielinczuk S., and Hofmann A. W. (1984) Hafnium/rare earth element fractionation in
the sedimentary system and crustal recycling into the Earth's mantle. Earth Planet. Sci. Lett. 69, 365-378.
Peucker-Ehrenbrink B., Hofmann A. W., and Hart S. R. (1994) Hydrothermal lead transfer from mantle to continental crust: the role
of metalliferous sediments. Earth Planet. Sci. Lett. 125, 129-142.
Pitcher W. S. (1979) Comments on the geological environment of
granites. In Origin of Granite Batholiths. Geochemical Evidence
(ed. M. P. Atherton and J. Tarney), pp. 1-8. Shiva.
Pohl J. R. and E m m e r m a n n R. ( 1991 ) Chemical composition of the
Sri Lanka Precambrian basement. In The Crystalline Crust of Sri
Lanka (ed. A. Kroener), Part 1; Geol. Surv. Dept. P r o f Paper5,
pp. 9 4 - 1 2 4 .
Poldervaart A. (1955) Chemistry of the Earth's crust. In Crust of the
Earth (ed. A. Poidervaart); G.S.A. Spec. Paper 62, 119-144.
Price N. B., Calvert S. E., and Jones P. G. W. (1970) The distribution
of iodine and bromine in the sediments of the South Western Barents Sea. J. Marine Res. 28, 2 2 - 3 3 .
Puchelt H. (1972) Handbook o f Geochemistry II, Sections B-O.
Springer-Verlag.
Rapp R. P. (1994) Partial melting of metabasalts at 2 - 7 GPa: Experimental results and implications for lower crustal and subduction zone processes. Mineral. Mag. 58A, 7 6 0 - 7 6 1 .
Ringwood A. E. (1975 ) Composition and Petrology of the Earth's
Mantle. McGraw-Hill.
1231
Ronov A. B. (1968) Organic carbon in sedimentary rocks. Geochimija 5, 4 0 9 - 4 2 3 .
Ronov A. B. and Yaroshevskiy A. A. (1968) Chemical structure of
the Earth's crust. Geochem. lntL 5, 1041-1066.
Ronov A. B., Grinenko V. A., Girin Y. P., Savina L. I., Kazakov
G. A., and Grinenko L. N. (1974) Effects of tectonic conditions
on the concentration and isotopic composition of sulfur in sediments. Geochem. Intl. 11, 1246-1272.
Rudnick R. L. (1992) Xenoliths-Samples of the lower continental
crust. In The Continental Lower Crust (ed. D. M. Fountain et al.),
pp. 269-316. Elsevier.
Rudnick R. L. and Presper T. (1990) Geochemistry of intermediate/
to high-pressure granulites. In Granulites and Crustal Evolution
(ed. D. Vielzeuf and P. Vidal), pp. 5 2 3 - 5 5 0 . Kluwer.
Rushmer T. ( 1991 ) Partial melting of two amphibolites: contrasting
experimental results under fluid-absent conditions. Contrib. Mineral. Petrol. 107, 4 1 - 5 9 .
Schermerhorn L. J. G. (1987) The Hercynian gabbro-tonalite-granite-leucogranite suite of Iberia: geochemistry and fractionation.
Geol. Rundschau 76, 137-145.
Schneider A. (1978) Handbook o f Geochemistry II, Sections 16,
C-E, M-O. Springer-Verlag.
Schwarcz H. P. (1969) Handbook of Geochemistry 11, Section 6, B.
Springer-Verlag.
Shaw D. M., Reilly G. A., M u y s s o n J. R., Pattenden G. E., and
Campbell F. E. ( 1 9 6 7 ) T h e chemical c o m p o s i t i o n of the Canadian Precambrian Shield. Canadian J. Earth Sci. 4, 8 2 9 854.
Shaw D. M., Dostal J., and Keays R. R. (1976) Additional estimates
of continental surface Precambrian shield composition in Canada.
Geochim. Cosmochim. Acta 40, 7 3 - 8 4 .
Shaw D. M., Cramer J. J., Higgins M. D., and Truscott M. G.
(1986) Composition of the Canadian Precambrian shield and
the continental crust of the earth. In The Nature o f the L o w e r
Continental Crust (ed. J. B. Dawson et al.), pp. 2 7 5 - 2 8 2 .
Blackwell.
Siewers U. (1974) Beitrag zur H~iufigkeit und Isotopenzusammensetzung des Schwefels in magmatischen und metamorpben Gesteinen. Dr. Dissertation, Univ. Gtittingen.
Sighinolfi G. P. (1973) Beryllium in deep-seated rocks. Geochim.
Cosmochim. Acta 37, 702-706.
Sighinolfi G. P. and Santos A. M. (1976) Geochemistry of gold in
Archean granulite facies terrains. Chem. Geol. 17, 113-123.
Sighinolfi G. P., Santos A. M., and Martinelli G. (1979) Determination of tellurium in geochemical materials by flameless atomic
absorption spectroscopy. Talanta 26, 143-145.
Smith J. D. and Burton J. D. (1972) The occurrence and distribution
of tin with special reference to marine environments. Geochim.
Cosmochim. Acta 36, 621-629.
Springer W. and Seck H. A. ( 1995 ) Partial fusion of basic granulites
at 5 to 15 kb. Implications for the origin of TTG magmas. Contrib.
Mineral. Petrol. (in press).
Stacey J. S. and K.ramers J. D. (1975) Approximation of terrestrial
lead isotope evolution by a two-stage model. Earth Planet. Sci.
Lett. 26, 207-221.
Sun S.-S., Nesbitt R. W., and Sharaskin A. Y. (1979) Geochemical
characteristics of mid-ocean ridge basalts. Earth Planet. Sci. Lett.
44, 119-138.
T a m e y J., Weaver B., and Drury S. A. (1979) Geochemistry of Archean trondhjemitic and tonalitic gneisses from Scotland and East
Greenland. In Trondhjemites, Dacites and Related Rocks (ed. F.
Barker), pp. 275-299. Elsevier.
Taylor S. R. and McLennan S. M. (1985) The Continental Crust: Its
Composition and Evolution. Blackwell.
Tepper J. H., Nelson B. K., Bergant W., and Irving A. J. (1993)
Petrology of the Chilliwack batholith North Cascades, Washington: generation of calc-alkaline granitoids by melting of lower
crust with variable water fugacity. Contrib. Mineral. Petrol. 113,
333-351.
Truscott M. G., Shaw D. M., and Cramer J. J. (1986) Boron abundance and localization in granulites and the lower continental crust.
Bull. Geol. Soc. Finland 58, 169-177.
1232
K . H . Wedepohl
Veizer J. and Hoefs J. (1976) The nature of O 18/O~6 and C FZ/CJ3
secular trends in sedimentary carbonate rocks. Geochim. Cosmochim. Acta 410, 1387-1395.
Voland B. (1969) Die Verteilung des Indiums in Eruptivgesteinen.
Ein Beitrag zur Geochemie des Indiums. Freiberger Forschungsh.
C246, 67-125.
Watson E. B. and Harrison T. M. (1983) Zircon saturation revisited:
temperature and composition effects in a variety of crustal magma
types. Earth Planet. Sci. Lett. 64, 295-304.
Weber J. N. (1960) Geochemistry of graywackes and shales. Science
131, 664-665.
Wedepohl K. H. ( 1969-1978) Handbook of Geochemistry. SpringerVerlag.
Wedepohl K. H. (1978) Handbook o f Geochemistry 11, Sections 73,
B-G. Springer-Verlag.
Wedepohl K. H. (1987) The chlorine and sulfur crustal c y c l e - Abundance of evaporites. In Geochemistry and Mineral Formation in the Earth Surface (ed. R. Rodriguez-Clemente and Y.
Tardy), pp. 3-27. Consejo Superior de Investigaciones Cientificas
Centre National de la Recherche Scientifique, Madrid.
Wedepohl K. H. ( 1991 ) Chemical composition and fractionation of
the continental crust. Geol. Rundsch. 80, 207-223.
Wedepohl K. H., Heinrichs H., and Bridgwater D. ( 1991 ) Chemical
characteristics and genesis of the quartz-feldspatic rocks in the
Archean crust of Greenland. Contrib. Mineral. Petrol. 107, 163179.
Whalen J. B., Currie K. L., and Chappell B. W. (1987) A-type granites: geochemical characteristics, discrimination and petrogenesis.
Contrib. Mineral. Petrol. 95, 407-419.
White W. M. (1993) 238U/2°4pb in MORB and open system evolution of the depleted mantle. Earth Planet. Sci. Lett. 115, 211 226.
Wlotzka F. (1972) Handbook of Geochemistry 11, Sections 7, B-M,
O. Springer-Verlag.
Wolf M. B. and Wyllie P. J. ( 1991 ) Dehydration-melting of solid
amphibolite at 10 kbar: textural development, liquid interconnectivity and application to the segregation of magmas. Mineral. Petrol 44, 151-179.
Ye S. (1992) Crustal structure beneath the central Swiss Alps derived
from seismic refraction data. Ph. D. thesis, ETH Ztirich.
Zartman R. E. and Haines S. M. (1988) The plumbotectonic model
for Pb isotopic systematics among major terrestrial reservoirs--A
case for bidirectional transport. Geochim. Cosmochim. Acta 52,
1327-1339.